There are more datasets that could be used for RM. Some of them are [HellaSwag](https://huggingface.co/datasets/hellaswag) [ELI5](https://huggingface.co/datasets/eli5) [Stanford Human Preferences Dataset](https://huggingface.co/datasets/stanfordnlp/SHP) [Open-Assistant dataset](https://huggingface.co/datasets/OpenAssistant/oasst1)